Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Mathematical Reasoning on AIME 2025 (Score, Best Score, ΔH)
Loading...
0.39
Score
RLOO
0.3068
0.3284
0.35
0.3716
Mar 12, 2026
Score
Best Score
ΔH
Updated 2mo ago
Evaluation Results
Method
Method
Links
Score
Best Score
ΔH
RLOO
Backbone=Qwen-3-32B, C...
2026.03
0.39
0.4
0.04
GRPO
Backbone=Qwen-3-32B, C...
2026.03
0.36
0.39
0.06
REPO-R
Backbone=Qwen-3-32B, C...
2026.03
0.35
0.38
0.17
ADAPO
Backbone=Qwen-3-32B, C...
2026.03
0.34
0.35
0.29
DAPO
Backbone=Qwen-3-32B, C...
2026.03
0.31
0.35
0.75
Feedback
Search any
task
Search any
task
